Related Experiment Video
Updated: Jun 13, 2025
The Motivation for Alcohol Reward: Predictors of Progressive-Ratio Intravenous Alcohol Self-Administration in Humans
Published on: April 28, 2022
You've got mail: comparing individuals who do and do not provide DNA saliva samples by mail in a study of severe
Michelle Eglovitch1, Alexis Edwards2, Kathryn Polak2
1Department of Psychology, Virginia Commonwealth University, 806 W. Franklin St., Richmond, VA 23284-2018, United States.
Objective:
Many genetic studies of psychiatric disorders rely on participants to mail in DNA samples. Differences in who returns a sample may affect the generalizability of these studies, but little attention has focused on possible differences between participants who do and do not provide samples. The present study compared participants with severe lifetime alcohol use disorder (AUD) who did and did not return saliva DNA samples.
Methods:
N = 3927 individuals were recruited through Facebook for a genome-wide association study of severe AUD. Participants completed an online survey and agreed to provide a saliva DNA sample by mail. Survey measures included: demographics, quantity and frequency of recent alcohol and tobacco use, impulsivity, and personality. Participants who returned the saliva kit (n = 2412) were compared to those who did not provide DNA samples (n = 1515) were compared using univariate Chi-square and t-tests.
Results:
The sample was predominantly White (88.3%) and female (64.0%). DNA providers were more likely than non-providers to report graduate-level education. DNA providers were also less likely to report recent cigarette smoking and scored higher on measures of conscientiousness, perseverance, and premeditation. Using multivariate regression, the most parsimonious model found that being male, a non-smoker, and endorsing perseverance was associated with providing DNA.
Conclusions:
In an online sample of individuals with severe AUD, participants who did not provide DNA samples showed multiple demographic and psychosocial differences from those who did provide samples. These findings may have implications for generalizability and suggest that further research is needed.
